Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add wan-huiyan/agent-traffic-control --skill subagent-reports-complete-but-pr-unmergedgit clone --depth 1 https://github.com/wan-huiyan/agent-traffic-controlWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-reports-complete-but-pr-unmerged)<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-reports-complete-but-pr-unmerged"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/subagent-reports-complete-but-pr-unmerged/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-reports-complete-but-pr-unmerged"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/subagent-reports-complete-but-pr-unmerged.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00372 | $0.02247 |
| Opus 5 | $0.00186 | $0.01123 |
| Sonnet 5 | $0.00074 | $0.00449 |
| Haiku 4.5 | $0.00037 | $0.00225 |
Grade A, and why
subagent-reports-complete-but-pr-unmerged scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sub-Agent Reports "Completed" While PR Is Still OPEN
Problem
You dispatch 3-5 parallel sub-agents to open + review + merge PRs that close GitHub issues. Each agent's brief explicitly says "wait for green CI, add auto-deploy label, squash-merge, delete branch, verify issue auto-closes." The agents work, push PRs, then return <task-notification status: completed>.
Several of those PRs are still OPEN. CI was still in-progress when the agent terminated. Or a reviewer found a P1 and the agent fixed it but stopped before re-running CI. Or the agent post-rebased and pushed but didn't merge. The parent, reading "completed," marks its TaskList task done. Hours later you realize 3 of 5 PRs never landed.
This is systematic, not occasional. In one session (scan-bugs-parallel, 2026-05-08, this repo) 4 of 4 sub-agents stopped before final merge:
- Wave1-G PR #527: agent posted P1 review, then stopped. Parent had to dispatch a continuation agent to fix + merge.
- Wave1-E PR #528: agent's final summary "Now I'll continue with the rest of the workflow (CI, auto-deploy label, merge)" — but the message ended there. PR was OPEN with green CI. Parent merged.
- Wave2 PR #540: agent merged successfully on its own (the exception). But it had to mid-flight rebase, which exposed the timing fragility.
- Wave2 PR #544: agent stopped after pushing the post-rebase branch but before CI completed. Parent merged.
Trigger Conditions
Apply this skill when:
- You've used
Agenttool withrun_in_background: trueto launch sub-agents whose briefs include "open PR + review + merge" as a single unit of work. - A
<task-notification status: completed>arrives, and:- The summary mentions "waiting for CI" / "pytest still pending" / "watching merge".
- The summary mentions "code review posted" without "merged".
- The summary mentions "rebased" / "pushed" without "merge SHA".
- You have a TaskList row keyed to GitHub issue closure that you're about to mark
completedpurely on the basis of the sub-agent notification. - You see a session-level total like "PRs merged: N" — and you got that count from sub-agent reports rather than
gh pr list --state merged.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 156 lines · 372 tokens per session scan A 3f36a338951a
subagent-reports-complete-but-pr-unmerged is a skill published in the GitHub repository wan-huiyan/agent-traffic-control (3 stars, last pushed today), licensed MIT. It adds 372 tokens to every session and 2,247 once invoked, about $0.0019 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
release
Cut a brooks-lint release: set the version in package.json, propagate it across all four plugin manifests and every version-bearing text file (README badges, docs site metadata), write the CHANGELOG entry, validate, then commit, push, tag, and publish the GitHub release. Triggers when the maintainer asks to "release"…
han-release
Cut a Han release: update CHANGELOG.md with the changes since the last release, bump and tag every plugin that changed as {plugin-name}--v{version} so a version-constrained dependency can resolve, and publish a GitHub release crediting every merged pull request and closed issue to the people behind it. Use when…
bump-sdk-version
Bump the FutureSearch SDK version across all files. Use when releasing a new SDK version, updating version numbers, or the user says bump version, release, version bump.
ship
Finish a branch: infer which review passes the change warrants, run them, open the PR, babysit CI to green, triage bot comments, and refresh the body from a clean context.
pr-check
PR review compliance: fetch review comments from an open PR, categorize as resolved/unresolved/dismissed, critically evaluate fixable items, implement approved fixes, and reply inline. Pass --unattended to halt on human-judgment items (emitted as Pending-Human) instead of prompting via AskUserQuestion.
babysit
PR babysitter: monitors CI status, auto-rebases when behind, auto-fixes CI where possible, delegates review comment handling to dlc:pr-check, and re-requests review after fixes. Designed for /loop usage with Remote Control.